What are your credentials, jcb? Have you ever worked in the industry? You certainly seem to think you know what you're talking about, but the experience?
[quote: altruistic.lemon]
As to whether you can hear a difference, isn't that as much to do with the speakers and the rest of the system as the resolution of the files, as some systems may not be able to accurately reproduce high res?
[/quote][quote: steve_1979]
The differences between 16/44.1 and 24/96 are less than -80dB down in volume or occur at ultrasonic frequencies over 20kHz. Even with the best hifi system in the world your ears won't hear it.
[/quote]
SOS Forum
[quote: Hugh Robjohns]
Again, few real world converters provide more than 20 bit performance anyway, and most domestic replay systems struggle to achieve the equivalent of 15 bits of dynamic range. The typical ambient background noise of a quiet domestic listening room might be 25SPL on a good day. Few high quality domestic speakers can produce peaks of 100dB SPL at the listening position. The very best with mighty amplification might achieve 110dB SPL or so -- if there are no neighbours or small children to worry about. So we're are looking -- at best -- of a usable dynamic range in the home of 75-85dB... which would require between 12 and 14 bits to encode perfectly, with the system noise floor swamped by the ambient acoustic noise floor, and sound level peaks painfully loud.
What I'm saying is that there is nothing wrong, restrictive, quality-limiting or inappropriate with 16 bit music reproduction systems for the domestic environment. It is actually the ideal format, in fact.
The 24 bit requirement comes from the need for higher quality (greater dynamic range and bandwidth) at the recording end of the chain so that the music production process has 'room to manoevre' -- so that unexpected peaks don't clip when recording, so that the noise floor doesn't become audible to the end user when dozens of tracks are mixed together, and so that the concatenation of band-limiting filters throughout the recording and post-production chain don't encroach on the wanted audio.
All of which means that 24 resolution is helpful at the recording and mixing end of the chain, but that 16 bit is absolutely fine for the reproduction end.
24 bit files offer no improvement in soundstage width, or mix dynamics, or timing fluidity... So if it sounded different it was a different mix. The only difference you should have heard would be a lower noise floor.
More information isn't recorded. This is a very common fallacy. A 16 bit system properly dithered can record and reproduce audio signals down to around -120dBFS the same as a 24 bit system. The only difference is that it's well into the noise floor in a 16 bit system, and roughly level with the noise floor in a 24 bit system. But it's the same audio information...
Look at it this way. Pros use 24 bit to enable a material to be recorded and mixed with adequate headroom without compromising the noise floor. Mastered tracks (currently) have no headroom. So to turn a 24 bit source recording to a mastered track you're going to introduce gain which raises the noise floor... And so you no long have 24 bit dynamic range any more. No 24 bit system delivers the theoretical 144dB signal-noise ratio. The best manage about 120dB, and most about 110dB. If a source recording leaves 15dB headroom -- which is not unusual -- removing the headroom to produce a masted track for commercial release will have a signal-noise ratio of 95dB at the very most... Which, funnily enough, is the same as a perfect 16 bit system.
As I said, 16 bits is entirely adequate and well optimised for consumer listening formats.
[/quote]
[quote: altruistic.lemon]
What are your credentials ... Have you ever worked in the industry? You certainly seem to think you know what you're talking about, but the experience?
[/quote]
LinkedIn
As to whether you can hear a difference, isn't that as much to do with the speakers and the rest of the system as the resolution of the files, as some systems may not be able to accurately reproduce high res?
[/quote][quote: steve_1979]
The differences between 16/44.1 and 24/96 are less than -80dB down in volume or occur at ultrasonic frequencies over 20kHz. Even with the best hifi system in the world your ears won't hear it.
[/quote]
SOS Forum
[quote: Hugh Robjohns]
Again, few real world converters provide more than 20 bit performance anyway, and most domestic replay systems struggle to achieve the equivalent of 15 bits of dynamic range. The typical ambient background noise of a quiet domestic listening room might be 25SPL on a good day. Few high quality domestic speakers can produce peaks of 100dB SPL at the listening position. The very best with mighty amplification might achieve 110dB SPL or so -- if there are no neighbours or small children to worry about. So we're are looking -- at best -- of a usable dynamic range in the home of 75-85dB... which would require between 12 and 14 bits to encode perfectly, with the system noise floor swamped by the ambient acoustic noise floor, and sound level peaks painfully loud.
What I'm saying is that there is nothing wrong, restrictive, quality-limiting or inappropriate with 16 bit music reproduction systems for the domestic environment. It is actually the ideal format, in fact.
The 24 bit requirement comes from the need for higher quality (greater dynamic range and bandwidth) at the recording end of the chain so that the music production process has 'room to manoevre' -- so that unexpected peaks don't clip when recording, so that the noise floor doesn't become audible to the end user when dozens of tracks are mixed together, and so that the concatenation of band-limiting filters throughout the recording and post-production chain don't encroach on the wanted audio.
All of which means that 24 resolution is helpful at the recording and mixing end of the chain, but that 16 bit is absolutely fine for the reproduction end.
24 bit files offer no improvement in soundstage width, or mix dynamics, or timing fluidity... So if it sounded different it was a different mix. The only difference you should have heard would be a lower noise floor.
More information isn't recorded. This is a very common fallacy. A 16 bit system properly dithered can record and reproduce audio signals down to around -120dBFS the same as a 24 bit system. The only difference is that it's well into the noise floor in a 16 bit system, and roughly level with the noise floor in a 24 bit system. But it's the same audio information...
Look at it this way. Pros use 24 bit to enable a material to be recorded and mixed with adequate headroom without compromising the noise floor. Mastered tracks (currently) have no headroom. So to turn a 24 bit source recording to a mastered track you're going to introduce gain which raises the noise floor... And so you no long have 24 bit dynamic range any more. No 24 bit system delivers the theoretical 144dB signal-noise ratio. The best manage about 120dB, and most about 110dB. If a source recording leaves 15dB headroom -- which is not unusual -- removing the headroom to produce a masted track for commercial release will have a signal-noise ratio of 95dB at the very most... Which, funnily enough, is the same as a perfect 16 bit system.
As I said, 16 bits is entirely adequate and well optimised for consumer listening formats.
[/quote]
[quote: altruistic.lemon]
What are your credentials ... Have you ever worked in the industry? You certainly seem to think you know what you're talking about, but the experience?
[/quote]
Claude Shannon, possibly one of the most brilliant mathematicians of all time, died at 84 with advanced Alzheimers.
But he was happy,
JC
But he was happy,
JC
busb
Well-known member
shadders said:
This has been answered several times! Without dither, the artefacts would be much higher than -90dB, so high, they'd often be audible. It's the added dither that "averages out" the quantisation error but at the expense of increasing the noise-floor upwards but at a level that's inaudible. It's one of those rare win-win situations!
These links, if you haven't already read they may help:
http://www.head-fi.org/t/415361/24bit-vs-16bit-the-myth-exploded
http://www.audiomisc.co.uk/inadither/Page1.html
busb
Well-known member
Jota180 said:
I agree with this post. As for AS, some think its the start of the human race's next evolutionary step. It may well not feel that way for you though. Someone close to me was diagnosed age 23, now 40. They suffer from high levels of anxiety.
Added:
I don't share your confidence in DB ABX testing - it doesn't deal with false negatives & needs to be conducted by professionals, is time consuming & rather boring to take part in so say nothing of the expense. No one in the industry is going to risk such tests, IMO. I've bee to demos of speaker cables where I've stated tha I couldn't really tell much difference to be challenged by the manufacturer with such statements as "Are you sure!" or "Well, it's pretty obvious to the rst of us!" etc. Such tests always start at the cheap end then work upwards. I never been to a demo whee the audiance has ever been asked to guess what level a particular cable falls in their range!
shadders
Well-known member
busb said:
Hi,
I had a quick scan - for the head-fi - they do state "perceived" increased S/N. Perceived is the key word here - not actual increase in S/N.
For the latter audiomisc - Jim Le Surf is using a single tone at -90dB - hence only using 2 bits from the possible 16bits - and i think he is using processor gain to achieve the 120dB S/N.
I would need to read more closely - to confirm the above.
Again, valid analysis, but the context does need to be explained, without, can be misleading.
busb said:
Can you explain how you have derived this statement - i was not able to locate in the text links you provided. Thanks.
Regards,
Shadders.
One reason why dither might be applied to a 16bit file, is if that file were created from a higher wordlength file (say 24bit), in which case truncation error would be present.
Dither would resolve that error.
Among other reasons might be decimation, or where a sample rate change has been involved.
Usually, it's when the file is the result from previous manipulations.
JC
Dither would resolve that error.
Among other reasons might be decimation, or where a sample rate change has been involved.
Usually, it's when the file is the result from previous manipulations.
JC
busb
Well-known member
shadders said:
A near full scale signal of say a pure sinewave is going to be encoded with thousands of levels. Reconstruction without dither is going to look very much like the original. If we encode the same sinewave but say -50dB below the 1st, this lower level sine is going to be encoded with far fewer levels. When reconstructed without dither, the waveform is going to be rather distorted. In other words, without dither, the level of distortion is going to be inversely proportional to amplitude. So, to preserve the purity of low level signals we either need to code with far more levels (such as 24bits instead of 16) OR we apply dither.
To keep things simple & within my ability to explain, we'll take dither just to be added white noise of a very low level. By adding this noise, we ensure that the encoding "breaks up" the the exact point of the repeating low level sinewave gets encoded. Instead of exactly the same point at a given phase angle being encoded indentically, we have effectively randomly offset the repetition in the time domain for each cycle. We have added a small amount of jitter to our low level sinewave but this instantaneous jitter gets averaged out. We are not modulating the sinewave with noise we are merely suming it with it which is itself is below audiblity. We have traded high distortion of low level signals for noise. That additional noise is inaudible so is why the trade-off works. You were unable to find the quoted sentence 'cos it wasn't a quote but my interpretation.
Pure sinewaves are used in most examples simply for illustration of the removal of harmonics. My attempted explanation isn't exact, avoids maths so only attempts to illustrate. TBH, Jim Lesurf does a far better job with the addition of diagrams. If my attempted explanation falls short, I encourage you to do what I do when I fail to understand something - re-read (his text) until the penny drops!
Dither should be added to any low-amplitude or highly-periodic signal before any quantization or re-quantization process, in order to de-correlate the quantization noise from the input signal and to prevent non-linear behavior (distortion); the lesser the bit depth, the greater the dither must be. The result of the process still yields distortion, but the distortion is of a random nature so the resulting noise is, effectively, de-correlated from the intended signal. Any bit-reduction process should add dither to the waveform before the reduction is performed.
In short, dither de-correlates quantisation errors from the signal source.
JC
In short, dither de-correlates quantisation errors from the signal source.
JC
busb
Well-known member
jcbrum said:
+1. Any unplanned deviation from the original is effectively distortion, be it f response, harmonic products, noise or any other undesired artefacts. JC's point regarding de-correlating the input signal from the quantisation process decouples the signal from that process. A damn elegant solution! The point that any downsampling needs to have dither (re)applied is very important, otherwise 24 to 16bit will sound worse for the wrong reasons not related to bit depth.
As an aside, my fairly new CA Azur 751BDR universal player will play back SACD discs despite owning none. The DDS stream gets downconverted to PCM to my external DAC so I need not worry about the sound from such discs being degraded - I would still be hearing the superior mastering (provided the player downsamples correctly). I'm on holiday today so am going to read up on noise shaping.
DSD is an interesting case in point when considering quantisation errors.
Since DSD has only a one bit wordlength, one might suppose that the quantisation error is huge. In fact so large as make one assume it might be insurmountable, since it is 32,000 times larger than 16bit.
The reality is that DSD is very heavily dependent on dither and, in particular, noise shaping technology, yet produces results which are audibly undetectably different from multibit technology, such as 24bit pcm.
JC
Since DSD has only a one bit wordlength, one might suppose that the quantisation error is huge. In fact so large as make one assume it might be insurmountable, since it is 32,000 times larger than 16bit.
The reality is that DSD is very heavily dependent on dither and, in particular, noise shaping technology, yet produces results which are audibly undetectably different from multibit technology, such as 24bit pcm.
JC
This white paper presented to the AES, gives details of a Comparitive Listening Test, between DSD and 24bit PCM.
JC
JC
p.s. Just in case any readers don't know what DSD is, the following note might be of assistance . . .
DSD is 1-bit, has a sampling rate of 2.8224 Mhz. The output from a DSD recorder alternates between levels representing 'on' and 'off' states, and is a binary signal (called a bitstream). The long-term average of this signal is proportional to the original signal. DSD makes use of noise shaping techniques in order to push quantisation noise up to inaudible ultrasonic frequencies. In principle, the retention of the bitstream in DSD allows the SACD player to use a basic (one bit) DAC design which incorporates a low-order analog filter. The SACD format is capable of delivering a dynamic range of 120 dB from 20 Hz to 20 kHz and an extended frequency response up to 100 kHz, although most currently available players list an upper limit of 80–90 kHz and 20 kHz is the upper limit of human hearing.
JC
DSD is 1-bit, has a sampling rate of 2.8224 Mhz. The output from a DSD recorder alternates between levels representing 'on' and 'off' states, and is a binary signal (called a bitstream). The long-term average of this signal is proportional to the original signal. DSD makes use of noise shaping techniques in order to push quantisation noise up to inaudible ultrasonic frequencies. In principle, the retention of the bitstream in DSD allows the SACD player to use a basic (one bit) DAC design which incorporates a low-order analog filter. The SACD format is capable of delivering a dynamic range of 120 dB from 20 Hz to 20 kHz and an extended frequency response up to 100 kHz, although most currently available players list an upper limit of 80–90 kHz and 20 kHz is the upper limit of human hearing.
JC
busb
Well-known member
altruistic.lemon said:
Oooh, a bullet whistling past from a sniper! What bits of digital technology you choose to believe doesn't change the technology. If it's boring you, why the hell post? If I feel sorry for anyone here, it certainly ain't me.
busb
Well-known member
jcbrum said:
Not everyone is that enamoured by DDS. There are some interesting fudges to single bit like multiple single bit (multi-level delta sigma modulators) Although there maybe some point to DDS, I'm not sure what the hell it is (in the context of ADCs & DACs) apart from extracting the analogue with very simple filtering I suppose. It has a very high noise figure but is spread over a huge BW (MHz).
My class D power amp uses single bit PWM that's similar to PFM (PDM).
Added:
By DDS, I mean DSD!
Alec
Well-known member
altruistic.lemon said:jcbrum said:
You seem a nice bloke, jcb, but only when you're not in patronising mode
And you're a really tedious bloke but only...hmmm, nope...
altruistic.lemon said:JC
You seem a nice bloke, jcb, but only when you're not in patronising mode
HaHa, same to you, with knobs on
JC
busb said:
Hehe, this always makes me laugh. Posters who join a thread simply to bajs on it. "It's boring", "Is this still going?", "just listen to the music", etc etc.
Fair enough first time, 1. See new thread, 2. Is it interesting to me? 3. Yes...contribute 4. No...Click on another thread and get on with life.
There are some people who contribute and some who spoil
There have been some great technical posts in this thread, and I am very glad I started it. Some great information (much over my head even if I have an A level in maths) and some interesting points.
The chread- trappers know who they are...they contribute nothing and are essentially trolling.
It's a big forum. There is lots to see. Lots to contribute. So why waste time if it is uninteresting? Seriously, I don't get it.
Facebook
Twitter
Instagram